1. Field of the Invention
The present invention is directed to a novel process for producing novel intermediates useful in the synthesis of 1-.beta.-alkyl carbapenems.
2. Description of the Prior Art
A wide variety of carbapenems, such as the natural fermentation product thienamycin (Formula I), have been reported in the patent and scientific literature as having exceptional antibacterial activity. ##STR4##
However, researchers attempting to develop thienamycin have encountered two problems, namely: (1) the compound is very difficult to ferment and isolate, and (2) the product is very unstable, such that it reacts with itself and decomposes. To circumvent these problems, carbapenem derivatives have been prepared which possess excellent stability and antibacterial spectra.
One such group of derivatives currently being investigated is the 1-.beta.-methyl carbapenems of the formula: ##STR5## wherein R.sup.1 is hydrogen or a conventional hydroxy-protecting group; and
R.sup.2 and R.sup.3 are independently selected from the group consisting of substituted and unsubstituted: alkyl, alkenyl and alkynyl, having from 1-10 carbon atoms; cycloalkyl, cycloalkylalkyl and alkylcycloalkyl, having 3-6 carbon atoms in the cycloalkyl ring and 1-6 carbon atoms in the alkyl moieties; spirocycloalkyl having 3-6 carbon atoms; phenyl; aralkyl, aralkenyl and aralkynyl wherein the aryl moiety is phenyl and the aliphatic portion has 1-6 carbon atoms; heteroaryl, heteroaralkyl, heterocyclyl and heterocyclylalkyl wherein the hetero atom or atoms in the above-named heterocyclic moieties are selected from the group consisting of 1-4 oxygen, nitrogen and sulfur atoms and the alkyl moieties associated with said heterocyclic moieties have 1-6 carbon atoms; wherein the substituent or substituents relative to the above-named radicals are selected from the group consisting of: amino, mono-, di- and trialkylamino, hydroxyl, alkoxyl, mercapto, alkylthio, phenylthio, sulfamoyl, amidino, guanidino, nitro, chloro, bromo, fluoro, cyano and carboxy; and wherein the alkyl moieties of the above-recited substituents have 1-6 carbon atoms. PA0 R.sup.2 is a lower alkyl having from 1-6 carbon atoms, PA0 R.sup.3 is hydrogen or a triorganosilyl group, and PA0 R.sup.4 represents a group of the formula ##STR10## wherein R.sup.5 and R.sup.6 independently, or taken together, are selected from the group consisting of hydrogen, substituted and unsubstituted: alkyl, alkenyl and alkynyl, having from 1-10 carbon atoms; cycloalkyl, cycloalkylalkyl and alkylcycloalkyl, having 3-6 carbon atoms in the cycloalkyl ring and 1-6 carbon atoms in the alkyl moieties; spirocycloalkyl having 3-6 carbon atoms; phenyl; aralkyl, aralkenyl and aralkynyl wherein the aryl moiety is phenyl and the aliphatic portion has 1-6 carbon atoms; heteroaryl, heteroaralkyl, heterocyclyl and heterocyclylalkyl wherein the hetero atom or atoms in the above-named heterocyclic moieties are selected from the group consisting of 1-4 oxygen, nitrogen and sulfur atoms, providing that when R.sup.5 and R.sup.6 are taken together to form said heterocycylic moiety, said moiety contains at least one hetero nitrogen atom, and the alkyl moieties associated with said heterocyclic moieties have 1-6 carbon atoms; wherein the substituent or substituents relative to the above-named radicals are selected from the group consisting of: amino, mono-, di- and trialkylamino, hydroxyl, alkoxyl, mercapto, alkylthio, phenylthio, sulfamoyl, amidino, guanidino, nitro, chloro, bromo, fluoro, cyano and carboxy; and wherein the alkyl moieties of the above-recited substituents have 1-6 carbon atoms, and PA0 which process comprises the steps of: PA0 (A) reacting a compound of the formula ##STR11## wherein R.sup.2 and R.sup.4 are as defined above, PA0 (B) reacting Compound 8 with a compound of the formula ##STR13## wherein R.sup.1 and R.sup.3 are as defined above, and
Recently reported synthetic schemes for producing 1-.beta.-methyl carbapenems of Formula II, such as those of Shih et al., Heterocycles, volume 21, no. 1, pages 29-40 (1984), proceed through intermediates of the formula ##STR6## from which the 1-.beta.-methyl carbapenems can be formed easily and in high yield. Unfortunately, however, these schemes require numerous other intermediates and time consuming steps to produce the above intermediate, each of which increases the process time and decreases the overall yield. Furthermore, the steps required to produce .beta.-methyl intermediate III also produce a large amount of the corresponding .alpha.-methyl product. Accordingly, there is a need for a stereoselective process which provides a high percentage .beta.-yield (the yield of .beta.-product/the total yield of product) of intermediates which can easily be converted to 1-.beta.-alkyl carbapenems.
One recent process of interest described by Tajima et al, in Tetrahedron Letters, volume 26, no. 5, pp. 673-676 (1985), discloses the reaction of silyl enol ether intermediates with 4-acetoxyazetidinone to produce compounds of the formula ##STR7## wherein R.sup.1 is a hydroxy-protecting group, and R.sup.2 is selected from the group consisting of ##STR8## wherein PNZ is p-nitrobenzyloxycarbonyl, and TMS is trimethylsilyl.
This procedure has been viewed with interest because it bypasses several steps of conventional carbapenem syntheses. However the interest has been limited because practioners now wish to produce 1-.beta.-alkyl carbapenem intermediates, and Tajima et al provides no guidance for a stereoselective process which provides a high percentage .beta.-yield of such intermediates.
Other procedures, such as those disclosed in U.S. patent application Ser. No. 725,594 filed Apr. 22, 1985, and now U.S. Pat. No. 4,683,296 patented July 28, 1987, which is a continuation-in-part of U.S. patent application Ser. No. 472,443 filed Mar. 7, 1983, use silyl enol ether precursors for preparing 1-.beta.-alkyl carbapenem intermediates. These processes, however, do not use azetidinone thiolesters and thus provide no guidance for processes which do.
Accordingly, it would be desireable to produce azetidinone thiolester intermediates by a silyl enol ether process, which intermediates can be used to produce 1-.beta.-alkyl carbapenems. Furthermore, it would be desireable that such process be stereoselective and provide a high percentage .beta.-yield.